Thunder & Lightnings

Gloster Javelin - History

Design model
A design proposal model similar to the P.280, showing most of the Javelin's features; author

The Javelin's story begins similarly to that of many RAF aircraft - a succession of requirements accompanied by a succession of ever-changing designs to match. In 1947, the Air Ministry had seen the need for a high performance interceptor to challenge the increasingly modern bomber designs being produced. Two separate aircraft were envisaged; one for day and one for night fighting. While the day fighter eventually led to the Hunter, the night/all-weather fighter requirement was down to a fight between Gloster and de Havilland (the latter submitting their DH.110, later to become the Sea Vixen). The specification was numbered F.44/46, and three each of the de Havilland DH.110 and the Gloster GA.5 were ordered. The F.44/46 specification matured to become F.4/48, covered by operational requirement OR.227, which called for a fighter capable of 525 knots at 40,000 ft, armed with 4 30mm cannon and 4 AAMs with a powerful radar.

Javelin prototype
WD804, the first prototype; Gloster

Gloster settled on a design (based on proposal P280) for a huge delta-winged aircraft with two Rolls-Royce Avon engines before rising weight forced them into selecting higher thrust Metrovick F.9 engines (later known as the Armstrong-Siddeley Sapphire) instead. In the meantime, a further two DH.110s were ordered, but the order for Gloster's submission was cut to two, the thinking being that the DH.110 was more likely to succeed. Gloster were not overjoyed at this; two prototypes were going to make slow going if the aircraft was chosen for service, and in 1951 a change of thinking at the Air Ministry finally persuaded them to order three more GA.5s. Later in the year the first prototype flew, on 26th November. WD804 was the aircraft in question and was a bare shell - no radar, no weapons. It had had a prolonged series of taxi trials, throwing up problem after problem, and the first flight was no different - with heavy vibration setting in soon after take off.

Second prototype
WD808, the 2nd prototype (bulge above the tail is the spin recovery parachute housing), 1952

The vibration problems were caused by the interaction of engine exhaust and airflow over the rear fuselage which required several redesigns of this area to finally fix. Sqn Ldr Bill Waterton (Gloster's chief test pilot), who had been with the project from the very first sketches, had already been critical of the insufficient fuel load and poor cockpit visibility, and the low hydraulic boost provided to the flying controls. As the litany of issues with the Javelin's design continued, Waterton tendered his resignation, fed up with the risks he was taking and the inertia of the company when it came to rectifying obvious problems - but they successfully convinced him to stay.

Then, on 2nd June 1952, Waterton took WD804 up for a trial flight in preparation an appearance at RAF Fighter Command's annual tactical conference. Having faced a succession of unpleasant flights in this aircraft already, Waterton wanted to be confident the jet would behave itself in front of an audience. The flight progressed smoothly until he was on his return to base, winding up the speed for a fast run. A couple of small thumps preceeded two and half seconds of chaos, with the aircraft shaking and buzzing and suddenly pitching down, with no response to pulling back on the stick. At high speed, Waterton knew an ejection was unlikely to be successful - and so, with the ground looming ever larger in his windscreen, he tried the pitch trimmer mechanism (which altered the angle of the entire high T-tail). It worked, and the jet pulled out of what would have been a fatal dive. Waterton stabilised her at 10,000 ft and 300 knots and made a mayday call, intending to take the aircraft out over the Bristol Channel and eject.

However, within a few minutes he was toying with the idea of getting the jet back down on the ground and saving this valuable prototype - and the experimental data held on the flight data recorder. He soon found he had just enough control to make this a possibility, and gingerly flew the aircraft to Boscombe Down's enormous runway. It all went wrong as the aircraft touched down - understandably, a little fast, as he had not dared use full flap, not knowing for sure how much damage there was behind his seat. The aircraft bounced, and a succession of higher bounces and harder arrivals back on the ground eventually resulted in the port gear leg being forced through the wing, tearing open a fuel tank and igniting a fierce fire. Initially trapped in the cockpit, Waterton managed to get the jammed canopy open and sprinted away from the flames. He then returned, directing the fire crew to pour water and foam over the nose, to save the flight data recorder. Waterton's feat was recognised with the awarding of the George Medal, but it was nearly the end of his patience with the aircraft - and the company.

The first production order had arrived, and with it the name of Javelin. The second prototype, WD808, flew in August but then stayed on the ground until January 1953 while research went on to determine the cause of the elevator loss on the first prototype. Tests throughout the first half of the year resulted in a number of changes to the design, including a cranked wing and beefed-up fuselage, fin and control surfaces. Waterton flew both the second and third prototypes, neither of which - in his view - performed any better, and were continuing to suffer from issue, with his reports and feedback continuing to be apparently ignored.

Fourth prototype
WT830, the 4th prototype (compare the cranked wing with WD804), 1954; Gloster

It was at this point that the Javelin's high T-tail cost the life of pilot Peter Lawrence - such designs are susceptible to a condition known as a deep stall, where the wing blankets airflow over the tail assembly at high angles of attack, rendering the aircraft uncontrollable. At some point in the flight the nose had been pulled past 45 degrees with flaps down, and as the stall came on, forward airspeed reduced to zero - the aircraft simply fell out of the sky. Lawrence fought for recovery all the way down, ejecting only at the last moment - too late to seperate from the seat. In March 1953 a third prototype (WT827) flew, being the first one to carry guns and radar. A number of differing radome designs were tried before they settled on the familiar pointed design. Large belly tanks were also fitted, as it was now finally agreed that the Javelin was lacking on fuel.

FAW.1 and FAW.2
FAW.1 XA620 & FAW.2 XA776 of 46 Squadron, 1957

A fourth prototype (WT830) flew in January 1954 and though lacking guns and radar, it did incorporate all the improvements made to the ill-fated second prototype, including powered ailerons. Waterton was tasked with delivering the jet to its flight test base, only to be told just before take-off that the new ailerons had been manufactured to much lower stress tolerances than planned, and would have failed had he exceeded 200 knots. He made the delivery flight - slowly! - but it was another illustration of the often needless perils of test flying that he was continuing to undergo. WT803 was soon passed to the A&AEE at Boscombe Down to be assessed by their pilots. At the end of March, Waterton had had enough and put his complaints together in a report to the senior men at Glosters, asking for them to be recognised or he would quit. Days later, he was out of a job, having been given a single day to clear his desk.

The fifth and final prototype, WT836, joined the flying programme in July, the same month in which one of the earlier prototypes was flown at through the sound barrier over London, the ensuing sonic boom causing a hell of a lot of fuss. Gloster's official position was that this was an accident, but it was widely believed to be a demonstration that the aircraft could indeed go supersonic - much press criticism, including from Bill Waterton (who was by now a newspaper correspondent and had made public some of his criticisms of the aircraft), had said it could not.

T.3 prototype
WT841, the T.3 prototype

While possible export customers settled down to watch the progress of the programme before expressing any real interest (and soon losing any they had), by mid-1956 over 20 FAW.Mk.1s were engaged in trials work - the type of development batch arrangement later formalised for production of later types such as the Lightning. Despite the designation of FAW.1, none of these aircraft were ready for real service - while 46 and later 87 Squadrons got used to their challenging new charges, they were covered by many limitations on the flight envelope and they were not missile armed. The FAW.2 introduced a new American radar (the APQ.43, instead of the FAW.1's ancient AI.17), with a larger radome, and went into service with 46 and later 89 Squadrons. The next variant was a trainer, the T.3, with tandem seating to avoid major changes to the design, all-moving tailplane but no radar. The change in centre of gravity meant a small fuselage extension was required, and this was used to house extra fuel (and make the aircraft substantially better looking in the process!). 228 OCU at Leeming received most of the T.3s but they were also spread around the various Javelin squadrons.

85 Squadron FAW.6s (XA832 closest camera); MoD

The FAW.4 was similar to the FAW.1 but had the T.3's all-moving tailplane and vortex generators on the wings and entered service with 141 Squadron at RAF Horsham St. Faith, and later with 3, 11, 23, 41, 72 and 87 Squadrons (though not all simultaneously). The FAW.5 was externally similar to the FAW.4 but had a redesigned wing interior in order to make room for more fuel, and provision for the full total of four Firestreak AAMs - though in the end the mark never carried them. The wing improvement of the FAW.5 was soon applied to the FAW.2, thus producing the FAW.6.

FAW.9 plans
FAW.9 plans

In November 1956 the FAW.7, the first mark to actually carry the four missiles specified in the original requirement, first flew. This was basically an FAW.5 but with uprated Sapphire Sa.7 engines and powered rudder and extended rear fuselage. By this time so many different marks of the Javelin were in the air it was a wonder anybody had any idea what was happening. A larger number of each mark were being used in various trials on such basic items as the weapons and engine fits, leading one to believe the Air Ministry had handed the RAF over to Gloster as one big Guinea Pig. The FAW.8, an FAW.7 with reheat, appeared so quickly that of the 142 Mk.7s produced, only 42 entered service and only these were fitted to carry missiles - the remaining 80 FAW.7s were instead hurried out of the factory doors and delivered straight into storage at RAF Kemble, later being converted to FAW.9s.

FAW.9 above Leuchars
29 Sqn FAW.9 above Leuchars, 1962; the late Brian Carroll's collection
(pilot - Brian, navigator - the late John Douglas Boyd)

It was not until June 1960 that an RAF Javelin - an FAW.7 - finally fired a Firestreak missile, successfully downing a Meteor drone. The reheat-capable FAW.8 was limited to using reheat only at a minimum altitude; below that point engaging the reheat actually caused a loss of thrust (to the point where take-off could not be safely accomplished with reheat engaged). This was down to the engine's fuel pump - it fed fuel at a constant rate and only at high altitude was there sufficient excess capacity to allow fuel to be burned directly without causing a loss of cold thrust at the same time. However the FAW.8 did have an improved, drooped, wing leading edge and autostabiliser to improve handling. The FAW.9 was basically an FAW.7 incorporating the FAW.8 improvements, and even then there was a couple more variations -the FAW.9F/R which could be equipped with a fantastically ugly and massive refuelling probe (obviously designed by somebody who took the name of the aircraft a little too literally) and the FAW.9R - R for Range, which could carry 4 underwing droptanks rather than the 2 that other Mk.9s could carry.

In service the Javelin had settled down to do a steady, unspectacular job of guarding the nation against the expected Soviet bomber fleets - and while designed as a medium range bomber destroyer, and subject to many restrictions on how it was flown, could put up a creditable performance against other aircraft of the same time - no contest against the upstart Lightning, but a fairly even match for a Hunter if flown carefully. Ironically, before the endless changes and versions had produced a useable Javelin fighter, it had been light enough on its feet that it could outfight the DH.110 - but in their final forms, the Sea Vixen could outperform the Javelin, having none of the latter's handling restrictions. However, it did lack guns so by the time it came to a close-in dogfight, the Sea Vixen pilot would have been helpless to do anything other than try the rarely used rocket pack, or hurl abuse over the radio.

The Lightning replaced the Javelin in the UK and Germany in short order with most being gone by 1965, but the Javelin held on for a few years longer in the Far East, where it gained its only air to air victory - an Indonesian C-130 which crashed while trying to avoid a Javelin that had been sent to intercept it during the Malayan crisis in 1964. However, with the increasing success of the Lightning, the Javelin's days even in hotter climes were numbered, and the last Javelin squadron was 60 Squadron, disbanding at RAF Tengah on Singapore, at the end of April 1968.

FAW.9s at Kai Tak
60 Sqn FAW.9s (XH893 foreground) during a deployment to Kai Tak, 1967

Ironically, while the Javelin had been preferred over the DH.110 because it was considered to offer more development potential, all of the really interesting proposals fell by the wayside and were never to see daylight. These included reconnaissance variants with extended noses and/or a greater wingspan, an attack version to carry bombs in under-fuselage panniers and most significantly a supersonic version with area-ruled fuselage, thinner wings and redesigned tail unit. This was being seriously explored when it was cancelled, and soon after Duncan Sandys' infamous Defence White Paper was published, cancelling most advanced aircraft projects. Gloster spent the years between this point and their absorption by BAC desperately trying to get back into the game, but never succeeded, and the Javelin was the final production type that Gloster produced.

FAW.9 climbing into the evening
FAW.9 climbing out of Leuchars, 1962; the late Brian Carroll's collection

The Javelin's protracted development period and lack of opportunity to prove itself in combat have led to it being described in scathing terms by many people, and its reputation was not helped by logistical and servicing mishaps in deployments overseas, but it was well liked by its pilots who appreciated the amount of weaponry available to them (far exceeding other types of the day), its stability (within the proscribed guidelines) and its roomy cockpit (despite poor view down to the sides - particular for the navigator). Any shortcomings it had in dogfighting ability were more than outweighed by its ability to stop the fight before the proverbial 'knife fight' began, and its airbrakes were incredibly effective - enough to force an attacker to overshoot before they realised what was happening, and often used to permit impressively steep descents to landing.

A single Javelin continued to fly with the RAE until 1976, when it was delivered to the Imperial War Museum's care at Duxford airfield, where it remains to this day. Lacking in any real popularity with the public, the large production run has sadly not been reflected in numbers of preserved examples, and a mere 10 complete Javelins now remain in various states of preservation, none of which will ever fly again.

Leading Particulars

First flight 26 Nov 1951 (prototype)
22 Jul 1954 (production)
31 Oct 1955 (proto)
25 Apr 1956 (prod)
20 Aug 1956 (proto)
6 Jan 1958 (prod)
19 Sep 1955 (proto)
27 Feb 1956 (prod)
26 Jul 1956 (prot)
24 Aug 1956 (prod)
15 Jan 1957 (proto) 9 Nov 1956 (prod) 9 May 1958 (prod) 1959?
Crew Two
Armament Four 30mm Aden cannon Two 30mm Aden cannon plus four Firestreak AAMs
Powerplant Two 8,000 lb Sapphire Sa6 Two 11,000 lb Sapphire Sa7 Two 11,000 lb (12,300 lb with reheat) Sapphire Sa7R
Max. speed 616 knots at sea level, 0.94 mach at altitude 555 knots at sea level, 0.91 mach at altitude 610 knots at sea level, 0.96 mach at altitude 612 knots at sea level, 0.93 mach at altitude 616 knots at sea level, 0.95 mach at altitude 610 knots at sea level, 0.93 mach at altitude
Service ceiling 52,500 ft 46,000 ft 50,700 ft 50,100 ft 52,800 ft 52,000 ft
Range ?
Max. take off weight 36,690 lb 37,200 lb 42,000 lb 37,480 lb 39,370 lb 40,600 lb 40,270 lb 42,510 lb 43,165 lb
Wing span 52 ft 10 in
Wing area 927 sq ft
Length 56 ft 3 in 55 ft 2 in 59 ft 11 in 56 ft 3 in 55 ft 2 in 56 ft 3 in 55 ft 2 in 56 ft 3 in
Height 16 ft 0 in
Production (total 435) 47 (including 7 prototypes) 30 22 50 64 33 142 47 116 (all FAW.7 conversions)

Production totals for each variant include converted airframes. Corrections and gap-filling welcome as always!

Externally the various variants can be easily confused, though there are a number of points to look for. The T.3 is easy to spot, with a tiny black radome. The FAW.1/4/5 and 2/6 are tricky to tell apart! The three flavours of FAW.9, FAW.9(F/R) and FAW.9R are particularly tricky to distinguish; if you see any drop tanks under the wings, it's an FAW.9R. If it has the refuelling probe fitted it's either a FAW.9(F/R) or 9R (and if the probe isn't fitted, there are two raised attachment lugs aft of the cockpit). But if has neither tanks nor probe... that doesn't necessarily mean it's definitely a plain FAW.9!

CanopyPitot probePylonsRear fuselageJetpipes
FAW.1LongNoNormalPort wingNoPen nibNormal
FAW.2ShortSometimesNormalPort wingNoPen nibNormal
T.3LongNoBulgedFirst 10 port wing, later bothNoPen nibNormal
FAW.4LongNoNormalPort wingNoPen nibNormal
FAW.5LongYesNormalPort wingNoPen nibNormal
FAW.6ShortYesNormalPort wingNoPen nibNormal
FAW.7LongNoNormalPort wingYesFlatNormal
FAW.8ShortNoNormalBoth wingsYesFlatExtended
FAW.9LongNoNormalBoth wingsYesFlatExtended